Menu
Case Studies
DEVELOPED A SOLUTION TO FORECAST DEMAND WITH HIGHER ACCURACY FOR A GLOBAL FERTILIZER COMPANY
A game-changing global fertilizer-producing company headquartered in India
The legacy demand forecasting process for the client’s global operation was Microsoft Excel-based and heavily dependent on personal judgment. This resulted in a low accuracy of 20-30%. Due to this, the client was facing a challenge in managing the inventory and placing replenishment orders from point of sale (PoS).
We followed the Prepare-Explore-Model process:
- Built a time series model to analyze the past data and forecast 6 months ahead
- External factors impact was introduced in the process
- The new system suggested a point forecast and four range forecast for each month in the horizon
- The forecast was now generated in a form that can be straight away entered into the Inventory Management Systems (IMS)
Our solutions enabled the client to:
- Overall forecast accuracy improved by 50%
- Reduced time is taken to generate forecasts
- Optimized inventory planning led to higher inventory turns to free up working capital
RESOURCE OPTIMIZATION SOLUTION TO IMPROVE UTILIZATION OF AIRPORT STAFF FOR AN AVIATION SOFTWARE COMPANY
A leading aviation software company headquartered in India
Increase resource utilization of airport staff
Built optimization models using Integer Linear Programming to optimize various resources needed for airport operations, considering various constraints like working hours, weekly-offs, holidays, overlapping shifts, etc.
- Models churned out optimum resources needed per shift, per day, per week.
- Generic models reduced code customizations needed for various airports across various countries.
- Improved resource utilization
- Faster and more accurate roster generation
DEVELOPED A MODEL TO PREDICT WHEAT CROP YIELD FOR A GOVERNMENT AGENCY
AGRICULTURAL RESEARCH INSTITUTE
- Obtain accurate crop yield prediction at the village level
- Crop yield prediction is a vital input for a lot of government policies, initiatives, and planning exercises. Traditional yield prediction methods are time-consuming and resource-intensive.
Used advanced machine learning techniques to improve the accuracy of crop yield prediction. Remote sensing data was used along with historical yield data and weather data.
- Improved crop yield prediction at the village level
- Reduced dependence on crop cutting experiments leading to huge savings in manpower cost and time
DEVELOPED ROBOTIC ADVISORY TO IMPROVE POLICY RECOMMENDATIONS FOR A GENERAL INSURANCE COMPANY
A general insurance company based in India
Subjectivity in recommending policies and coverage
- Built recommendation engine to propose policy and coverage for new or renewal customers.
- Engine to scan past data and analyze which policies & coverage are sold to similar customers.
- Perform gap analysis to identify up-selling or cross-selling opportunities.
- Brought more objectivity to policy & coverages recommendations, reducing dependence on human intervention.
- GAP Analysis led to the sale of more coverages.
- Increased customer satisfaction because of accurate policy & coverage recommendations
DEVELOPED AN ANALYTICS ENGINE FOR A SMART HOME AUTOMATION COMPANY
A leading smart-home automation company based in India
Data generated by IoT (Internet of Things) sensors was gathered but was not analyzed for insights or recommendations.
- Visualizing usage patterns, recommending actions, and setting up reminders
- Analytics data-mart built to store data on hourly, daily, weekly, monthly basis.
- Visualizations providing insights about the usage pattern of home appliances.
- Actions suggested to set up alarms & corrective measures.
- Improved insights leading to improved customer engagement with the application
HR TRAINING METRICS TO IMPROVE EMPLOYEE ENGAGEMENT FOR AN ANALYTICS COMPANY
A Revenue Optimization Company located in India
HR department did not have any tool to view and measure the 360-degree effectiveness of training programs.
- Built data-mart and visualization engine to give a 360-degree view of data.
- Created an Analytics engine that calculated training metrics, drainage, and suggested corrective measures.
Improved employee engagement and reduced training drainage
- Improvement in the engagement of employees & cost savings from reduced training leakage.
